Elon Musk's AI firm blames unauthorised change for chatbot's rant about 'white genocide'

[BumRushDaShow]

Source: The Guardian

Fri 16 May 2025 06.32 EDT
Last modified on Fri 16 May 2025 07.29 EDT


Elon Musk’s artificial intelligence company has blamed an “unauthorised modification” for a glitch in its Grok chatbot that resulted in the tool ranting about “white genocide” in South Africa. In a post on Musk’s X platform, xAI said new measures would be brought in to ensure its employees cannot modify the bot’s behaviour without extra oversight.

The Grok bot repeatedly referred to a white genocide in South Africa – a discredited claim promoted by Donald Trump among other US populist figures – this week in responses to unrelated queries. One user of Musk’s X platform, which also hosts Grok, asked the bot to identify the location of a photo of a walking path, triggering a non sequitur swerve into “South Africa’s farm attack debate”.

xAI, the Musk-owned company that developed the chatbot, said in a post on X that the bot’s erratic behaviour was due to an unauthorised change made to the Grok bot’s system prompt, which guides a chatbot’s responses and actions. “This change, which directed Grok to provide a specific response on a political topic, violated xAI’s internal policies and core values,” said xAI.

Indicating that a rogue employee was to blame, the post added new measures would be brought in to ensure that xAI employees “can’t modify the prompt without review”. It said the code review process for prompt changes had been “circumvented” in the incident. A new 24/7 monitoring team is also being brought in to deal with answers not caught by automated systems, xAI said. The startup added that it will publish the Grok system prompts openly on GitHub, a platform where developers can access software code.

Read more: https://www.theguardian.com/technology/2025/may/16/elon-musks-ai-firm-blames-unauthorised-change-for-chatbots-rant-about-white-genocide

---

"It's easier to beg forgiveness than ask permission."

The desired intention has now been put out there.

[Lovie777]

1. Come on............

admit it musk, you are a white supremacist like the rest of the shithole republicans.

musk spouts it like a badge of honor.

[Eugene]

2. I don't believe Musk, but it's a scary situation if he's telling the Truth.

An AI based on unreviewed crowd-sourced development is trusted to make decisions. Emerald Boy wants his AI to run the government.

[LiberalArkie]

20. More than likely training itself on its Lord and Master's Twitter messages.

[BOSSHOG]

3. All the men who pretend to be manly

Spend hours a day blaming others for whatever. Real men ask permission.

[mdbl]

7. They all have podcasts which makes up the manosphere

Which continues the Fux Nooze brain-washings.

[BoRaGard]

4. Fox 'investigation' finds "Biden To Blame"

as always. G.O.P.'s corporate propaganda pump is reliable for BEARING FALSE WITNESS

[Ray Bruns]

5. Why do I get the feeling that Grok is nothing more than Elon Musk sitting at a computer

and answering information requests?

[JT45242]

6. Apparently having the AI quote your CEO is now a glitch

Perhaps if they didn't use the words of apartheid Clyde, the company CEO, in their corpus it wouldn't spout Nazi propaganda

[travelingthrulife]

9. This is why AI is such a joke and will be a huge mess

Whoever puts the data in controls the message, as we see here.

[Attilatheblond]

15. AI presently seems like 'garbage in/garbage out' combined with Joe Isuzu

And does anybody wonder if Elon has a multiple personality disorder among all the other mental disorders he has?

[travelingthrulife]

8. ALL of his changes are unauthorized.

[WokeAintWhatIAm]

10. Hmmm...

Does rogue employee's name rhyme with,

Can't be Trust?

[Geechie]

11. The GOOD part!

https://www.facebook.com/photo.php?fbid=999654545677949]

The act of veering off into descriptions of “white genocide” was in effect completely demolishing the premise. The bot did more harm to Elon Musk than anyone else.

[Geechie]

12. More from the own-goal files:

https://www.facebook.com/photo.php?fbid=999654499011287]

[ToxMarz]

13. A rogue employee was to blame

The company has taken measures to ensure this won't happen in the future by promoting said employee to a senior managment postion where his duties will no longer involve "optimising" the AI code.

[eggplant]

14. The "rogue employee" thing is a red herring.

It is that the system is so easily corrupted that is the fundamental problem. This sort of thing is just going to continue.

[LudwigPastorius]

16. Better headline: Nazi Builds Nazi AI

“Grok mit uns” eh Elon?

[Martin68]

17. Translation: "unauthorised modification" means "we made a big mistake."

[area51]

18. "unauthorised change"

Eloon owns both twitter and xAI; he no doubt has to sign off on changes.

[HipChick]

19. This definitely got the okey dokey

from someone that had the access to implement

[BidenRocks]

21. Magats favorite word

BLAME!
Never accept responsibility.